The present invention relates to biosensors and chemical sensors. More particularly, it relates to sensors having a chemical or biochemical species detection group connected to an electronic circuit by electrically conducting polymer strands.
Biosensors employing enzymes have been applied to the detection of numerous analyte species concentrations including glucose, cholesterol, or both glucose and cholesterol concentrations in whole blood samples. Such sensors and associated instruments employ an enzyme capable of catalyzing, a reaction at a rate representative of the selected compound concentration in an assay mixture.
There are three general detection approaches employing a glucose enzyme electrode. The first and earliest measures oxygen consumption. The oxygen-sensing probe is an electrolytic cell with a gold (or platinum) cathode separated from a tubular silver anode by an epoxy casting. The anode is electrically connected to the cathode by electrolytic gel, and the entire chemical system is isolated from the environment by a thin gas-permeable membrane (often Teflon). A potential of approximately 0.8V (from solid-state power supply) is applied between the electrodes. The oxygen in the sample diffuses through the membrane and is reduced at the cathode with the formation of the oxidation product, silver oxide, at the silver anode. The resultant current is proportional to the amount of oxygen reduced. The analyzer unit operates over the range from 0.2 to 50 ppm of dissolved oxygen. Gases that reduce at 0.8V will interfere; these include the halogens and SO.sub.2. H.sub.2 S contaminates the electrodes.
A second approach detects H.sub.2 O.sub.2 production but requires an applied potential of approximately 0.65V (from solid-state power supply) applied between the electrodes, one of which is inside a permselective membrane. The H.sub.2 O.sub.2 in the sample diffuses through the permselective membrane (if one is present) and is oxidized at the anode. Many metal, metal complexes, nonmetal, organic and biochemical species that oxidize at approximately 0.65V will interfere; such as ascorbic acid, amines, hydrazines, thiol compounds, catechols, hydroquinones, ferrocenes, and metalloporphyrins. The inside permselective membrane is not always capable of removing the complicated mix of possible interferences from the analyte matrix.
A third approach takes advantage of the fact that the enzymatic reaction requires two steps. First, the enzyme glucose oxidase (GOD) (EC 1.1.3.4) is reduced by glucose, then the reduced enzyme is oxidized to its initial form by an electron acceptor, i.e., a mediator. In natural systems, the mediator is oxygen. In biosensors, another mediator compound may be employed to transfer electrons between the enzyme and a conductive surface of an electrode at a rate representative of the enzyme catalyzed reaction rate when an appropriate potential is applied to the particular redox mediator in use. Such biosensors may employ amperometric measurements to determine glucose concentration in a whole blood sample. This involves an integrated sample measurement of the area under the ampere versus time curve, corresponding to the amount of glucose in the sample.
The mechanism by which a common amperometric sensor works is depicted in FIG. 1. A sensor 2 employs glucose oxidase (GOD), for example, as a molecular recognition group. Glucose oxidase catalyzes the oxidation of glucose to gluconolactone in analyte 4. This reaction involves the FAD/FADH.sub.2 redox center of the enzyme. Sensor 2 includes a molecular recognition group, region 6, attached to an electrode 8. When glucose in analyte 4 contacts GOD-FAD (glucose oxidase including the FAD redox center) in region 6, it is oxidized to gluconolactone. At the same time, the GOD-FAD is reduced to GOD-FADH.sub.2. This involves two electrons and two hydrogen ions being transferred to the FAD. Normally, in the absence of a sensor mediator, the GOD-FADH.sub.2 is reoxidized by atmospheric oxygen to GOD-FAD to complete the catalytic reaction. In the presence of a mediator, however, the GOD-FADH.sub.2 is sometimes reoxidized by a mediator (M.sub.ox). In this case, the GOD-FADH.sub.2 releases two hydrogen ions to analyte 4 and two electrons to the mediator. The resulting reduced mediator (M.sub.red) may then be reoxidized by electrode 8 at an appropriate potential. The reoxidation of the mediator is accompanied by the transfer of an electron or electrons to electrode 8. This is the current that is monitored to provide a concentration of glucose.
In theory, a mediator may be any small molecule inorganic, organometallic or organic compounds, which are reduced by the enzyme, and oxidized by an appropriate applied potential at the electrode surface. The mediator should be designed to rapidly and efficiently transfer electrons between the enzyme and the electrode. Otherwise, ambient oxygen would oxidize nearly all of the reduced GOD and the desired signal would be very weak. The mediator should also transfer a total charge proportional to the glucose or cholesterol concentration in the sample. The current which results from the mediator oxidation is known as the Cottrell current which, when integrated with respect to time, gives the number of coulombs associated with the sensor reaction. The total coulombs passed is proportional to the amount of analyte.
Unfortunately, mediators are commonly provided as mobile "reagents" which diffuse to the enzyme where they are oxidized or reduced (depending upon the reaction catalyzed by the enzyme). The oxidized or reduced mediator then diffuses to the electrode surface where it gains or loses an electron. Unfortunately, such mechanism is dependent upon the continuing presence of recycled mobile mediators. As such compounds can leak from the electrode surfaces, there may be a gradual depletion in available mediator and a consequent reduction in sensor sensitivity. Examples of diffusing redox mediators include dyes (e.g., methylene blue), ferrocene derivatives (Cass, A E G; Davis, G; Francis, G D; Hill, H A O; Aston, W J; Higgins, I J; Plotkin, E V; Scott, L D L; Turner, A P F: Ferrocene-Mediated Enzyme Electrode for Amperometric Determination of Glucose. Anal. Chem. 56:667-671, 1984), components of conducting organic metals and quinones.
Also, available sensors applying the above amperometric approach to the detection of glucose, cholesterol, lactate, H.sub.2 O.sub.2, NAD(P)H, alcohol, and a variety of other compounds in whole blood samples, can have other serious complicating problems. For example, the percentage of sensor surface area covered by blood can vary; sometimes the blood sample does not cover the entire electrode. This may be caused by a poorly adherent enzyme (often applied by spraying) thus allowing leakage of blood or other analytes along the edges of the electrode. A related problem results from hydration of the reaction area prior to test. This dilutes the ligand (e.g., glucose) concentration and therefore gives a lower reading than would be accurately given by an unhydrated surface.
Further, the partial pressure of molecular oxygen (O.sub.2) may complicate the interpretation of sensor data. Molecular oxygen is the natural electron acceptor mediator of the enzyme glucose oxidase (GOD). Following oxidation of D-(+)-glucose by GOD, reduced glucose oxidase (GOD.sub.red) will transfer electrons to O.sub.2 forming H.sub.2 O.sub.2 in the absence of other mediators. In amperometric glucose biosensors described above, the unwanted O.sub.2 side reaction competes with synthetic chemical mediators for electrons supplied by the GOD.sub.red enzyme. Calibration of GOD-based biosensors at different altitudes (i.e., different partial pressures of O.sub.2) may be a problem if electron transfer rates of selected synthetic chemical mediators are not orders of magnitude faster than the O.sub.2 side reaction.
Humidity (i.e., H.sub.2 O) may be another potential problem if mass action of H.sub.2 O and O.sub.2 present drives the enzyme catalyzed oxidation product of D-gluconolactone in reverse back to the reduced starting material, D-(+)-glucose. Catalase, a common contaminant of glucose oxidase preparations, may be driven in reverse by mass action of excess H.sub.2 O and O.sub.2 producing 2 moles of H.sub.2 O.sub.2. H.sub.2 O.sub.2 buildup combined with D-gluconolactone could drive the glucose oxidase reaction in reverse by mass action back to D-(+)-glucose.
Other problems associated with known amperometric sensors include, for example, (1) difficulty in fitting the Cottrell current curve (i.e., ampere-time graph), (2) sampling with enough frequency to accurately obtain the time integral of Cottrell current, (3) high applied potential at the electrode causing indiscriminate oxidation or reduction of interfering substances, and (4) complicated electronic circuits requiring potentiostat and galvinostat instrumentation.
Some of the above drawbacks of the current amperometric biosensors have been noted and analyzed (see, Schuhmann, W: Chap. 9. Conducting Polymers And Their Application In Amperometric Biosensors. In: Diagnostic Biosensor Polymers. ACS Symposium Series 556. Usmani, A M; Akmal, N; eds. American Chemical Society; Washington, D.C.; 1994; pp. 110-123). First, due to the fact that the active site of redox enzymes is in general deeply buried within the protein shell, direct electron transfer between enzymes and electrode surfaces is rarely encountered. This is especially true for enzymes which are integrated within non-conducting polymer membranes in front of the electrode surface. Hence, electron transfer is usually performed according to a `shuttle` mechanism involving free-diffusing electron-transferring redox species for example the natural electron acceptor O.sub.2 or artificial redox mediators like ferrocene derivatives (Cass, A E G; Davis, G; Francis, G D; Hill, H A O; Aston, W J; Higgins, I J; Plotkin, E V; Scott, L D L; Turner, A P F: Ferrocene-Mediated Enzyme Electrode for Amperometric Determination of Glucose. Anal. Chem. 56:667-671, 1984), osmium complexes (Heller, A: Electrical Wiring of Redox Enzymes. Acc. Chem. Res. 23(5):128-134, 1990), or quinones. Due to the necessity for the redox mediators to diffuse freely between the active sites of the enzymes and the electrode surface, these electrodes show a limited long-term stability as a consequence of the unavoidable leaking of the mediator from the sensor surface. Additionally in the case of the natural redox couple O.sub.2 /H.sub.2 O.sub.2, the sensor signal is dependent on the O.sub.2 partial pressure, and a high operation potential has to be applied to the working electrode giving rise to possible interferences from cooxidizable compounds. The second drawback is related to the fabrication of these sensors. The physical assembling of an enzyme membrane and an electrode is extremely difficult to automate and thus in principal incompatible with microelectronic fabrication techniques. Additionally, the miniaturization as well as the integration of individual biosensors into a miniaturized sensor array is impossible with techniques which are mainly based on the manual deposition of a droplet of the membrane-forming mixture onto the electrode surface.
Consequently, the next generation of amperometric enzyme electrodes has to be based on immobilization techniques which are compatible with microelectronic mass-production processes and easy to miniaturize. Additionally, the integration of all necessary sensor components on the surface of the electrode has to prevent the leaking of enzymes and mediators simultaneously improving the electron-transfer pathway from the active site of the enzyme to the electrode surface.
In addition to amperolnetric mechanisms, which rely on detecting current generated from faradaic reactions, a potentiometric mechanism may be employed to sense analyte concentration. Potentiometric techniques monitor potential changes between a working electrode and a reference electrode in response to charged ion species generated from enzyme reactions on the working electrode. A very common potentiometric sensor is the pH sensor which registers changes in hydrogen ion concentration in an analyte. A microelectronic potentiometric biosensor, the Field Effect Transistor (FET) biosensor, has generated some interest. In this design, a receptor or molecular recognition species is coated on a transistor gate. When a ligand binds with the receptor, the gate electrode potential shifts, thereby controlling the current flowing through the FET. This current is detected by a circuit which converts it to an observed ligand concentration. Observed problems with potentiometric systems include, for example, (1) slow response of the electrode (i.e., seconds), (2) complicated electronic circuits for three electrode (i.e., working, counter, and reference electrode) electrochemical systems requiring potentiostat instrumentation, (3) low sensitivity, and (4) limited dynamic range.
Recently, two groups (Heller et al. and Skotheim et al.) have explored and developed redox polymers that can shuttle electrons from the enzyme to the electrode. The groups have "wired" the enzyme to the electrode with a long redox polymer having a dense array of electron relays. Each relay is a redox site bound to the polymer backbone. Electrons move along the polymer by hopping from one redox appendage to the next. The polymer penetrates and binds the enzymes, and is also bound to the electrode.
Heller et al. have conducted work on Os-containing redox polymers. They have synthesized a large number of such Os-containing polymers and evaluated their electrochemical characteristics (Gregg, B A; Heller, A: Redox Polymer Films Containing Enzymes. 1. A Redox-Conducting Epoxy Cement: Synthesis, Characterization, and Electrocatalytic Oxidation of Hydroquinone. J. Phys. Chem. 95:5970-5975, 1991). Their most stable and reproducible redox polymer is a poly(4-vinyl pyridine) to which Os(bpy).sub.2 Cl.sub.2 has been attached to 1/6th of the pendant pyridine groups. The resultant redox polymer is water insoluble. To make it water soluble and biologically compatible, Heller et al. have partially quatemized the remaining pyridine pendants with 2-bromoethyl amine. The redox polymer is water soluble and the newly introduced amine groups can react with a water soluble epoxy e.g., polyethylene glycol diglycidyl ether and GOD to produce a cross-linked biosensor coating-film. Such coating-films produced high current densities and a linear response to glucose up to 600 mg/dL (U.S. Pat. No. 5,262,035 to Gregg et al.).
Heller describes the electrical wiring of redox enzymes for use as amperometric biosensors (Heller, A: Electrical Wiring of Redox Enzymes. Acc. Chem. Res. 23(5):128-134, 1990). The Heller approach is an improvement over amperometric enzyme electrodes based on diffusing redox mediators, including dyes, ferrocene derivatives, components of conducting organic metals, and quinones, all described above. In the Heller approach, redox centers of a redox polymer polycation (e.g., 2[Os-(2,2'-bipyridine).sub.2 (poly(vinylpyridine))Cl].sup.1+/2+) are electrostatically and covalently bound to the enzyme and relays electrons to the electrode, on which a segment of the polycation is adsorbed. Binding of the redox polymer polycation to the electrode can be electrostatic when the electrode has a negative surface charge.
Fluctuations in current with partial pressure of oxygen (e.g., oxygen concentration in blood), depend on the ratio of the rate of direct electroxidation of the FADH.sub.2 centers to their rate of oxidation by molecular oxygen, and therefore on the rate of electron transfer to, and the electrical resistance of, the three-dimensional wired-enzyme structure. At high osmium-complex concentrations, and in sufficiently thin layers, the competition is won by electron transfer to the electrode via the osmium centers, and the electrodes are relatively insensitive to oxygen (Heller, A: Electrical Wiring of Redox Enzymes. Acc. Chem. Res. 23(5):128-134, 1990. Gregg, B A; Heller, A: Cross-Linked Redox Gels Containing Glucose Oxidase for Amperometric Biosensor Applications. Anal. Chem. 62:258-263, 1990. Surridge, N A; Diebold, E R; Chang, J; Neudeck, G W: Chap 5. Electron-Transport Rates In An Enzyme Electrode For Glucose. In: Diagnostic Biosensor Polymers. ACS Symposium Series 556. Usmani, A M; Akmal, N; eds. American Chemical Society; Washington, D.C.; 1994; pp. 47-70).
Electrodes based on conducting polypyrroles with ferrocenes also have been reported (Hale, P D; Inagaki, T; Karan, H I; Okamoto, Y; Skotheim, T A: A New Class of Amperometric Biosensor Incorporating a Polymeric Electron-Transfer Mediator. J. Am. Chem. Soc. 111(9):3482-3484, 1989).
Skotheim et al. have used flexible polymer chains to act as relays. Their polymers provide communication between GOD's redox centers and electrode. No mediation was found when ferrocene was attached to a non-silicone backbone. Their ferrocene-modified siloxane polymers were said to be stable and non-diffusing (Boguslavsky, L I; Hale, P D; Skotheim, T A; Karan, H I; Lee, H S; Okamoto, Y: Novel Biosensors For Specific Neurotransmitters Based On Flavoenzymes And Flexible Redox Polymers. Polym. Mater. Sci. Eng. 64:322-323, 1991).
Unfortunately, the redox polymer systems of Heller et al. and Skotheim et al. have a limited electron transfer rate based on electron hopping between dense electron relay pendant groups. Further, their "wire" redox centers must be designed to undergo reaction at a potential close to that of the enzyme catalyzed reaction. The closer the potential is to the redox potential of the enzyme itself, the lesser the likelihood that a potentially interfering substrate will be spuriously oxidized. Unfortunately, to address this issue limits the range of polymer redox couple and molecular headgroup combinations.
A fundamental presupposition for the construction of reagentless amperometric enzyme electrodes is the design of a suitable electron-transfer pathway from the active site of the enzyme to the electrode surface. According to Marcus theory (Marcus, R A; Sutin, N: Electron Transfers In Chemistry And Biology. Biochim. Biophys. Acta 811:265-322, 1985) a redox mediator with a low reorganization energy after the electron transfer has to be able to penetrate into the active site of the enzyme to shorten the distance between the prosthetic group (e.g., FAD/FADH.sub.2) and the mediator. Hence, the rate constant of the electron-transfer reaction can be increased. After this `first` electron transfer the redox equivalents have to be transported to the electrode surface via mechanism having a rate constant which is in the range of the turnover rate of the enzyme. In the shuttle mechanism mentioned above (having mobile mediators), the electron transport involves diffusion of redox mediators. In the "wired" redox polymer sensors described above, electron transport involves hopping from one redox center to the next on the polymer backbone.
In a recent study, Aizawa et al. discuss a reversible electron transfer between the prosthetic group of pyrrolo quinoline quinone (PQQ) enzyme (fructose dehydrogenase) and an electrode through a molecular interface (Aizawa, M; Khan, G F; Kobatake, E; Haruyama, T; Ikariyama, Y: Chap. 26. Molecular Interfacing of Enzymes on the Electrode Surface. In: Interfacial Design and Chemical Sensing. ACS Symposium Series 561. Mallouk, T E; Harrison, D J; eds. American Chemical Society, Washington, D.C., 1994, pp.305-313). The PQQ moieties of randomly oriented fructose dehydrogenase (FDH) which are very close to the transducer electrode can easily transfer their electrons to the electrode (Shinohara, H; Khan, G F; Ikariyama, Y; Aizawa, M: Electrochemical Oxidation and Reduction of PQQ Using a Conducting Polypyrrole-Coated Electrode. J. Electroanal. Chem. 304:75-84, 1991. Khan, G F; Shinohara, H; Ikariyama, y; Aizawa, M: Electrochemical Behaviour of Monolayer Quinoprotein Adsorbed on the Electrode Surface. J. Electroanal Chem. 315:263-273, 1991). However, the prosthetic groups of FDH located far from the electrode can not provide their electrons, as the distance from the electrode exceeds the maximum electron transfer distance (.about.25 .ANG.). Therefore, to make the FDH (EC 1.1.99.11, MW: 141,000) on the electrode surface electrochemically active, Aizawa et al. introduced an ultrathin conductive polypyrrole (PP) membrane as a molecular interface as "wiring" to assist the electron transfer from PQQ to the electrode. Unfortunately, the wiring used by Aizawa is randomly oriented and does not necessarily present enzyme at optimal position with respect to the analyte.
What is needed is an improved sensor design that rapidly transfers electrons from headgroup redox reactions to an electrode, does not rely on a redox relay such as freely diffusing mediators, and optimally orients the headgroup with respect to the analyte.
A great number of approaches for microfabrication of chemical sensors are currently under way, particularly in the areas of field effect transistor (FET)-based chemical sensors, metal oxide gas sensors, and biosensors. Since Janata et al. first reported micro-enzyme electrodes based on FET (Caras, S; Janata, J: Field Effect Transistor Sensitive to Penicillin. Anal. Chem. 52:1935-1937, 1980), a number of groups have been employing microfabrication techniques (e.g., photolithography) such as those employed in semiconductor device technology to fabricate micro-enzyme electrodes. Despite enormous efforts of many groups, the FET-based micro-enzyme electrodes of practical use have not been realized yet, largely because of the problems associated with potentiometric methods general lack of a fast response, high sensitivity, and wide dynamic range.
For the construction of reagentless enzyme electrodes (e.g., electrodes analogous to those of Heller et al. and Aizawa et al.) one has to focus on a technique for the modification and functionalization of electrode and even micro-electrode surfaces to allow the strong binding of the enzyme and the redox mediator taking into account the presuppositions for an effective and fast electron transfer between the enzyme and the electrode. These features requirements are in principle met with enzyme electrodes based on redox-sensitive hydrogels, however, the manual deposition of these hydrogels is not compatible with mass-production techniques.
The electrochemical deposition of conducting-polymer layers occurs exclusively on the electrode surface and can hence be used for the immobilization of enzymes either covalently using functionalities on the polymer film or physically entrapped within the growing polymer film. As the conducting-polymer film itself does not participate in the electron transfer, mediator-modified enzymes entrapped within a polypyrrole layer have been used for the construction of a reagentless oxidase electrode.
Electrochemical deposition methods of the prior art typically use high current density and voltage potential conditions which destroy the orderly Helmholtz double-layer at the electrode surface (U.S. Pat. No. 5,215,631 to Westfall). Resulting disorderly depositions at electrode surfaces produce random polymer structures which lack orientational and positional order. Aizawa et al. "wired" PQQ-FDH in their sensors with ultrathin conductive polypyrrole (PP) membrane as a molecular interface. Electrochemical synthesis of molecular-interfaced FDH on Pt electrode was prepared by the following two steps: (1) potential-controlled adsorption of FDH, and (2) electrochemical polymerization of polypyrrole. These steps employ high voltage and current density electrochemical deposition conditions to produce polymer (FDH and polypyrrole) depositions on the Pt electrode that are randomly oriented. Therefore, this device must operate at high (.about.400 mV) operating potential resulting in possible interfering cooxidizable species.
What is needed is an improved technique for depositing molecular recognition groups and associated wiring, if necessary, that provides a strong direct connection between an electrode and the molecular recognition groups, and allows the molecular recognition groups to be aligned in a common orientation.